DESCRIPTION: (Applicant's Abstract) The hypothesis of the proposed studies is that modulation of prostatic oncogenes by androgen deprivation can augment the efficacy of oncogene-specific recombinant viral vaccine therapy of prostate cancer. This hypothesis will be tested by: 1) Combining androgen ablation and recombinant poxvirus vector vaccination in rat Dunning prostate cancer cells expressing a model target antigen (beta galactosidase); 2) Using a transgenic prostate cancer model (in which SV40 T antigen serves as an androgen-responsive, prostate specific oncogene) to hormonally control oncogene expression during direct tumor cell vaccination with vaccinia vectors encoding SV40 T antigen. Preliminary data have shown that gene-modified tumor cell vaccines are effective therapy in an animal model of prostate cancer. This strategy, however, has significant limitations to potential clinical application. First, cumbersome culture of prostate-derived cells, which may not conserve unique aspects of the tumor genotype, is required. Second, efficacy is limited to microscopic disease. Third, cellular vaccine therapy targets undefined tumor antigens. The proposed studies seek to resolve these issues by using recombinant poxvirus vaccines, encoding specific tumor antigens, to treat and prevent prostate cancer in the Dunning rat model expressing a model tumor antigen, and in transgenic murine prostate cancer induced by prostate-specific expression of SV40 T antigen transgene. To obviate any need for culture of vaccine cells, recombinant vaccinia vector encoding the target prostate cancer oncogene (SV40 T antigen) will be used for direct host vaccination. To augment recombinant vaccine efficacy in treating advanced tumor burden, castration will be used both to mitigate oncogene-specific tolerance and to reduce host tumor cell burden. If castration and vaccination with vaccinia encoding T antigen are insufficient to abrogate tolerance, vaccinia vectors encoding T antigen in tandem with immunostimulatory cytokine genes will be used for therapy. To optimize coordination of androgen withdrawal and tumor-antigen-encoding vector vaccination, a model system using beta galactosidase as a target antigen in androgen responsive Dunning tumors will be evaluated before studies targeting SV40 T antigen. Molecular and cellular studies will dissect mechanisms operational in these models. The durability of therapeutic effect will be tested by re-induction of oncogene expression via androgen restoration. These studies aim to provide a framework for the eventual development and clinical application of recombinant viral vaccines targeting human prostatic antigens in the setting of androgen deprivation therapy.